Using quote() and substitute() for conditions within a shiny app - r

Question

I'm trying to use quote/substitute as a method to apply a condition in shiny. I'm not very familiar with either quote/substitute or shiny - so it is definitely possible that I'm not going about this problem in the right way.
I've created a simple example below that I illustrates the problem I get.
#Create test dataframe
test<-data.frame(x=c(0:10), y=c(rep(1,5),rep(2,6)), z=c("A","A","A","B","B","B","C","C","C","C","C"))
#example of what I would like to do outside shiny app
test[test$x > 5,]
#or using quote and eval
test[eval(quote(test$x > 5)),]
All of the above code works. But now lets say I want to apply it within a shiny app (and allow the user to choose the condition):
#create simple shiny app
require(shiny)
# Server
server <- function(input, output) {
# subset of nodes
df <- reactive({
#eliminate certain observations
x <- test[eval(input$condition),]
})
output$table <- renderTable({
df <- df()
})
}
# UI
ui <- fluidPage(
radioButtons("conditon", "Condition", choices = c("cond_1" = substitute(test$x > 5), "cond_2" = substitute(test$x<5))),
tableOutput("table")
)
# Create app
shinyApp(ui = ui, server = server)
But this gives the error "All sub-lists in "choices" must be names"). I'm not sure how to interpret this, and so am stuck. I looked at the answers in Shiny - All sub-lists in "choices" must be named? but did not find them helpful.
Would appreciate a way to solve this, or suggestions of a better approach (though note that I can't create the subsets ahead of time, as for my more complex actual example this creates issues).

Answer 1

A quick fix could be to wrap with deparse and then use eval(parse. It is not entirely clear why the input needs to be expressions. If this is just for subsetting, there are easier ways to accomplish the same
library(shiny)
-ui
ui <- fluidPage(
radioButtons("conditon", "Condition",
choices = list(cond_1 = deparse(substitute(test$x > 5)),
cond_2 = deparse(substitute(test$x<5))),
selected = deparse(substitute(test$x > 5)) ),
tableOutput("table")
)
-server
server <- function(input, output) {
# subset of nodes
df <- reactive({
#eliminate certain observations
test[eval(parse(text=input$conditon)),, drop = FALSE]
})
output$table <- renderTable({
df()
})
}
-Create app
shinyApp(ui = ui, server = server)
-output